Hill-type muscle models are broadly applicable to the assessment of motor function and critical to the design of improved rehabilitative strategies, serving as key components of muscle- driven simulations aimed at identifying factors that limit mobility due to age or neuromuscular impairment. Despite the ubiquitous use of Hill-type models, few studies have examined their accuracy and validity under in vivo, time-varying conditions. An overarching goal of the proposed project is to test and refine methods for assessing human muscle function using advanced Hill-type models, together with non-invasive measures of electromyographic activity and muscle structure. The lead innovative aim is to implement our recently developed two-element Hill-type model, with independent fast and slow contractile elements, within muscle-driven simulations of human cycling and to predict gastrocnemius forces across a range of speed and ergometry conditions that compare favorably to the forces determined experimentally from 3D ultrasound-based measures of tendon strain. Our previous work showed that a novel two-element model, driven by recruitment patterns of fast and slow motor units derived from EMG recordings, generates significantly better predictions of in situ and in vivo muscle force than traditional one-element models with average fiber properties. However, due to the large size of our goat animal model, we were unable to experimentally assess the muscles' F-V characteristics, which may have diminished the model's predictive capability. The proposed research addresses this limitation by using a new small animal model (rat distal hindlimb muscles) to conduct innovative in situ analyses of F-V and cyclical power output under varying stimulation conditions. These analyses, together with in vivo 3D X-ray imaging of muscle shape changes, will further advance the two-element model. The current work addresses two specific aims, critically broadening the impact of Hill-type models on clinical assessment of human motor function related to rehabilitation: Aim #1 evaluates the accuracy with which one- vs two-element models can estimate time-varying muscle forces within subject-specific simulations of human subjects pedaling on a cycle ergometer, using novel 3D ultrasound-based measures of tendon strain, fascicle pennation and muscle thickness. Aim #2 examines how motor unit recruitment and stimulation frequency affect in situ muscle mechanical output, with the goal of better predicting muscle force and power in situ and in vivo. Refinement of the two-element model will be based on in situ contractile dynamics, novel in vivo muscle-tendon force and fascicle strain measures, and innovative 3D X-ray video fluoroscopy. Insights from Aim 2 will be iteratively incorporated into the simulations of human cycling tested in Aim 1.